NASA Technical Reports Server (NTRS) 20000086133: Evolution of Tropical and Extratropical Precipitation Anomalies During the 1997 to 1999 ENSO Cycle PDF
Preview NASA Technical Reports Server (NTRS) 20000086133: Evolution of Tropical and Extratropical Precipitation Anomalies During the 1997 to 1999 ENSO Cycle
EvolutionofTropicalandExtratropicalPrecipitationAnomalies Duringthe1997to 1999ENSOCycle By ScottCurtis, JointCenterforEarthSystemsTechnology UniversityofMarylandBaltimoreCounty LaboratoryforAtmospheres NASA/Goddard Flight Center Code 912 Greenbelt, MD 20771 USA curtis @agnes, gsfc.nasa.gov phone: 301-614-6309 fax: 301-614-5492 Robert Adler, Laboratory for Atmospheres NASA/Goddard Flight Center George Huffman, Eric Nelkin, and David Bolvin Science Systems Applications Inc. Laboratory for Atmospheres NASA/Goddard Flight Center 97-98ENSO Abstract 1. Introduction 2. DataandIndices 3. EvolutionofPrecipitationin theTropics 3.1 Overview 3.2 Initiation of El Ni_o 3.3 Mature Phase of El Ni_o and Transition to La Ni_a 4. Global Patterns of Normalized Precipitation Anomalies 5. Observed Versus Modeled Precipitation 6. Conclusions Acknowledgements Figure Captions Figures References Abstract: The 1997-1999 ENSO period was very powerful, but also well observed. Multiple satellite rainfall estimates combined with gauge observations allow for a quantitative analysis of precipitation anomalies in the tropics and elsewhere accompanying the 1997-99 ENSO cycle. An examination of the evolution of the El Nifio and accompanying precipitation anomalies revealed that a dry Maritime Continent preceded the formation of positive SST anomalies in the eastern Pacific Ocean. 30-60 day oscillations in the winter of 1996/97 may have contributed to this lag relationship. Furthermore, westerly wind burst events may have maintained the drought over the Maritime Continent. The warming of the equatorial Pacific was then followed by an increase in convection. A rapid transition from E1 Nifio to La Nifia occurred in May 1998, but as early as October-November 1997 precipitation indices captured substantial changes in Pacific rainfall anomalies. The global precipitation patterns for this event were in good agreement with the strong consistent ENSO-related precipitation signals identified in earlier studies. Differences included a shift in precipitation anomalies over Africa during the 1997-98 E1 Nifio and unusually wet conditions over northeast Australia during the later stages of the El Nifio. Also, the typically wet region in the north tropical Pacific was mostly dry during the 1998-99 La Nifia. Reanalysis precipitation was compared to observations during this time period and substantial differences were noted. In particular, the model had a bias towards positive precipitation anomalies and the magnitudes of the anomalies in the equatorial Pacific were small compared to the observations. Also, the evolution of the precipitation field, including the drying of the Maritime Continentandeastwardprogressionof rainfall in theequatorialPacific, was lesspronouncedforthemodelcomparedtotheobservations. KEYWORDS: ENSO,ElNifio,LaNifia,Precipitation,Tropics 1. Introduction It canbearguedthatthe 1997-98E1Nifio wasthe strongestENSOeventever recorded. The unusuallywarmwatersof theequatorialPacific wereaccompaniedby changesinthelarge-scalecirculation,followedby precipitationanomaliesrangingfrom severedroughtto floods. The Maritime Continent,Amazon andCongobasins,and Central America experienceddrought during somepart of 1997 and 1998,while Argentina,Peru,andEastAfrica were hardest hit by flooding. In fact, there have been many studies documenting the variability of regional precipitation during the 97-98 event (Montroy et al. 1998, Bell and Halpert 1998, Bell et al. 1999, Jaksic 1998, Mullen 1998, Pavia and Badan 1998, Harrison and Larkin 1998, Kogan 1998, Jensen et al. 1998, McPhadden 1999, Kousky 2000). The El Nifio was followed by a persistent La Nifia. The regional precipitation anomalies for this phase of the ENSO cycle have been less well documented. Global precipitation during the 1997-99 ENSO was well observed from space. The SSM/I microwave and high resolution geo-IR data were available to monitor daily to monthly precipitation rates over even remote areas of the globe. The 1998-99 portion of the event was observed by the Tropical Rainfall Measuring Mission (TRMM) as discussed by Adler et al. (2000). In contrast, oceanic precipitation analyses for the comparable 1982-83 ENSO were limited to less capable satellite data, such as the outgoing longwave radiation information, or model data. With this motivation, satellite observations and gauges are used to examine the evolution of tropical and extratropical precipitation anomalies associated with the 1997-98 E1 Nifio and 1998-99 La Nifia. This work is intended to be a case study and the results found here may not be applicable to all 2 ENSO events. Finally, the observationswill be compared to model generated precipitationduringthisperiod. 2. Data Theprimary datasourcefor thisstudyisanexperimentalversionof themonthly GlobalPrecipitationClimatologyProject's(GPCP;Huffmanetal. 1997)communitydata set,hereafterreferredto asGPCP. Thisdatasetisdescribedby Huffmanetal. (1997), excepttherehasbeenaslightmodificationin themergertechnique;it usesTOVS-based estimatesofprecipitation(Susskindetal. 1997)tofill in missingor uncertaindatain the high latitudes,after being adjustedto GPCPvaluesin nearbyareas. The recordis extendedbackbeforetheSSM/I periodusingtheOLR PrecipitationIndex(OPI)of Xie andArkin (1998). The record of GPCP global monthly precipitation now extends from January 1979 to the present. The GPCP analysis has the distinctive feature of using the relatively infrequent low-orbit microwave observations to calibrate or adjust the geosynchronous IR-based estimates, thereby retaining the bias of the superior instantaneous rainfall observations and the superior sampling of the geosynchronous observations. GPCP is heavily weighted toward a gauge analysis over land (Rudolf et al. 1994). In general bimonthly averages of precipitation anomalies are shown, thereby reducing the effects of 30-60 day oscillations. However, individual months and twenty-- year climatologies were examined as well. GPCP is compared with NCEP/NCAR's reanalysis product (Kalnay et al. 1996). Precipitation is computed from a 62 wave triangular, 28 layer spectral model "Medium Range Forecast" run. Fortheexaminationof smallertime andspacescales,aonedegreedaily global precipitationdataset(1DD; Huffmanetal.2000)wasused. 1DDincludesprecipitation estimatesfrom geosynchonousIR, andTOVS. 1DDis computedseparatelyfrom the monthlyGPCPanalysis,buteachdayof thedailycombinationis scaledsuchthateach monthof daily datasumsto thecorrespondingGPCPmonthlyestimate.This is doneto ensureconsistencybetweenGPCPproductsandtointroducegaugedataasaconstraint on 1DD. The product currently extends from January 1997 to October 1999. NCEP/NCAR alsoproducesa daily reanalysisprecipitationdataset,with zonal grid spacingof 1.875°,andthis wasusedtocomparewiththedaily 1DDfields. 3. EvolutionofPrecipitationin theTropics 3.1 Overview This section presents a timeline of evolution of key components of the ocean- atmosphere system during the 1997-99 ENSO cycle. In particular, SST and precipitation anomalies are shown in Figs. 1 and 2. The beginning of 1997 saw a deep thermocline and warm ocean in the West Pacific and a shallow thermocline and cold ocean in the East Pacific (Bell and Halpert 1998). This was followed in early boreal spring by a rapid transition to E1 Nifio conditions in the ocean. Anomalous warming in the east was accompanied by a flattening of the thermocline. A drying of the Maritime Continent preceded the SST increase in the east-central Pacific and the following increase in precipitation there. The anomalously dry conditions over the Maritime Continent and 4 warm waters over the central Pacific peakedat the end of 1997. The maximum precipitationanomalyinthePacificfollowedinearly1998. AstheENSOstrengthenedc,hangeswereobservedin theprecipitationanomaly field in the centralPacific, which led the full onsetof La Nifia. A dipole of wet anomaliesalong the equatoranddry anomaliesto the north wasconsistentwith an intensifiedmeridional-verticalcirculation(CurtisandHastenrath1997)frommid-1997to mid-1998. The negativeprecipitation anomaliesin the Pacific precededthe positive anomaliesovertheMaritime Continent. This change-overin theatmospherecoincided with rapiddecreasesintheSSTandsubsurfacestructure(Belletal. 1999). The SST andprecipitation variability describedabove wasquantified using indicesovertheMaritime ContinentandPacificbasin.SST(ReynoldsandSmith 1995) and precipitation anomalieswere computed within a fixed box over the Maritime Continent(INDO; Fig.3). Nino 3.4(Fig.3)waschosentoquantifySSTanomaliesin the eastPacific,andprecipitationanomalieswerecalculatedforthisdomain. Becauseofthe varyingnatureof rainfall,movingboxeswerealsousedtodescribeprecipitationchanges. Areaaveragesof precipitation,thesizeof INDO andNino 3.4,weremovedthroughout largerdomainsencompassingtheMaritime continent (MC) andeastPacific(P)(Fig.3), asdescribedby CurtisandAdler(2000). HereMC-(+) denotestheminimum(maximum) value in the moving block averagesfound within MC and P-(+) the minimum '(maximum) value found within P. The same search procedure was used on anOmalous precipitation (aMC-, aMC+, aP-, aP+), which should be related to anomalies in the vertical velocity field associated with the components of the Walker Circulation. Because these indices search for the largest anomalies in an area, they can often lead 5 fixed-locationindicesin detectingregionalrainfallchanges(CurtisandAdler2000). An exampleofthesemovingboxindices,aP+andaMC-,isgiveninFig.4 (areproductionof Fig. 2 in Curtis andAdler (2000)) for monthsduring the 1997-98El Nifio. The normalizeddifferenceof aP+minusaMC- is theE1Nifio Index.(El), a measureof the westwardgradientofrainfall anomalies.ThenormalizeddifferenceofaMC+minusaP- is the La Nifia Index (LI), a measureof the opposite,eastwardgradientof rainfall anomalies.TheEl Nifio IndexminustheLaNifiaIndexyieldstheENSOPrecipitation Index(ESPI). TheEI,LI, andESPIarerelatedtotheanomalousstrengthoftheWalker Circulation,andarecorrelatedwell with traditionalENSOindices,suchasNino 3.4and theSouthernOscialltionIndex(SOl). Time seriesplotsfrom 1996to 1998areshownin Fig.5forNino 3.4(sst,precipitation),INDO (sst,precipitation),aMC+,aMC-,aP+,aP-, andFig.6forEI,LI, andESPI. 3.2 Initiation of El Ni_o The period of initiation and intensification of the El Nifio was roughly January to December 1997 (see bold lines in Figs. 5 and 6). Prior to that, the climate system was in the La Nifia phase. It was the fourth strongest La Nifia according to over 20 years of ESPI data, but relatively weak compared to the long-term SST and SO1 records. Early indications of a change in the climate system were seen through decreases in precipitation averages over the Maritime Continent (Fig. 5b,c) in September-October 1996. During November-December 1996 (Fig. la,b) the Maritime Continent was generally wet, and the central Pacific dry. However, an area of negative precipitation anomalies is found over Indonesia and Borneo, which is consistent with negative aMC- and INDO(precipitation) 6 valuesforthisperiod(Fig.5b,c). InJanuary-February1997(Fig.ld) thedryregionover the easternequatorialIndian OceanandIndonesiaintensified. Nino 3.4(sst)became positivein March 1997(Fig.5a). Infact,ofthestandardequatorialPacificSSTindices, onlyNino 4 (5 °N - 5 °S; 160°E - 150°W) showedaconcurrentevolutiontopositive valuesascomparedto aMC- andINDO(precipitation). TheMadden-JulianOscillation (MJO)mayhaveinfluencedthedecreasein precipitationover theMaritime Continent beforetheappearanceof El Nifio SSTs.Fig.7 showsaglobaltime-longitudediagramof daily precipitation averagedfrom 5° N to 5° S using the global, daily analysis of Huffmanetal. (2000)(seesection2). The periodis January1 1997to December31 1998.AnactiveMJOisapparentfrom 140Eto 160Wformostof 1997. InMarch-April 1997awarmanomalyemergedoff thecoastofPeru(Fig. le). In theEastPacifictheclimatologicalweakerbandof precipitationsouthof theequatorwas absent,whileintheWestPacifictheITCZ andSouthPacificConvergenceZone(SPCZ) werefurtherapartthannormal. Thispatternofprecipitationwasconsistentwithnegative precipitationanomaliesovertheMaritime Continent,extendingeastwardin abandalong theequatorandpositiverainfall anomalieslocatedimmediatelytothenorthandsouth (Fig. If). Theclearingalongtheequator(shownfromNovember1996toApril 1997) mayhavecontributed,throughincreasedshortwaveradiation,toanintensificationofthe warmingoftheequatorialcoldtonguein thiscase.Thisis consistentwithWeare(1983) whohypothesizedthatincreasedsurfaceheatingplays a role in the earlystagesof a developingE1Nifio. However,furtherinvestigationof therole ofcloudsintheinitiation of the1997-98E1Nifio is needed.In anyevent,warmE1Nifio watershadextendedJrJ_to the centralPacific by May-June 1997(Fig. lg). At the sametime the rain band
The list of books you might like
